This application is for a competing continuation of the currently funded project Internet Interviewing and the HRS. The current project is a joint collaboration between RAND and the University of Michigan. Its broad objective is to explore the use of Internet interviewing both as a possible substitute for the core interviews of the Health and Retirement Study and as a supplement to the core survey. In addition it aimed to improve the HRS survey instrument through a series of experiments on both HRS and non-HRS subjects. These experiments studied mode effects, selectivity, measurement of beliefs and preferences, measurement of health and health histories, visual presentation of tasks, and measurement of objective quantities such as wealth and consumption. Based on the success of this research, we propose in this application a continuation of this collaboration. Building on what we have learned, we propose additional experiments and investigations aimed at enhancing the HRS instrument, and more generally at improving measurement of core concepts in health, aging, and retirement. Examples of specific content are: measurement of preferences, expectations, cognition, health and well-being. In addition, we will develop and test a complete HRS core survey instrument that can be administered over the Internet, and which at some point could be substituted for the current face-to-face or telephone interview modes. We will also develop smaller Internet surveys that can substitute for the off-year surveys that HRS current undertakes, such as the Consumption and Activities Mail Survey. We will conduct these experiments and base these developments partly on Internet interviews with HRS subjects. To limit interview burden of HRS respondents and to avoid contamination of responses to the core survey, we will conduct a good deal of the research on the American Life Panel, a standing Internet panel of approximately 1500 subjects that has been developed at RAND.